Pub Date : 2024-08-02DOI: 10.1109/LSSC.2024.3437771
Muhammad Asfandyar Awan;Khalil Ahmad;Amine Bermak;Kabir H. Biswas;Bo Wang
This letter presents a photocurrent readout for bioluminescence detection. The design incorporates an asynchronous architecture employing a proposed capacitive feedback transimpedance amplifier (C-TIA) with a self-timed reset network and an all-digital reconfigurable time-domain quantization scheme. It eliminates the need for a periodic reset signal required in conventional C-TIAs and offers a wide dynamic range (DR) of 124 dB, a nonlinearity of 1.7%, and a 1-pArms input-referred noise while drawing only $210~mu $ A from a 1.8-V supply. Fabricated in a standard 180-nm CMOS technology, it occupies an area of 0.16 mm2. This design aims to facilitate in vitro NanoLuc (NLuc) luciferase-based bioluminescence sensing for biomolecule quantification at room temperature, with preliminary biological testing presented in this letter.
{"title":"An Asynchronous CMOS Current Readout With 124-dB Dynamic Range for Bioluminescence Sensing","authors":"Muhammad Asfandyar Awan;Khalil Ahmad;Amine Bermak;Kabir H. Biswas;Bo Wang","doi":"10.1109/LSSC.2024.3437771","DOIUrl":"https://doi.org/10.1109/LSSC.2024.3437771","url":null,"abstract":"This letter presents a photocurrent readout for bioluminescence detection. The design incorporates an asynchronous architecture employing a proposed capacitive feedback transimpedance amplifier (C-TIA) with a self-timed reset network and an all-digital reconfigurable time-domain quantization scheme. It eliminates the need for a periodic reset signal required in conventional C-TIAs and offers a wide dynamic range (DR) of 124 dB, a nonlinearity of 1.7%, and a 1-pArms input-referred noise while drawing only \u0000<inline-formula> <tex-math>$210~mu $ </tex-math></inline-formula>\u0000A from a 1.8-V supply. Fabricated in a standard 180-nm CMOS technology, it occupies an area of 0.16 mm2. This design aims to facilitate in vitro NanoLuc (NLuc) luciferase-based bioluminescence sensing for biomolecule quantification at room temperature, with preliminary biological testing presented in this letter.","PeriodicalId":13032,"journal":{"name":"IEEE Solid-State Circuits Letters","volume":null,"pages":null},"PeriodicalIF":2.2,"publicationDate":"2024-08-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141991494","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-07-25DOI: 10.1109/LSSC.2024.3433610
Abhishek A. Kadam;Shubham Patil;Ajay K. Singh;Maryam Shojaei Baghini;Udayan Ganguly;Laxmeesha Somappa
In advanced high-speed integrated systems, the widely distributed and proliferation of temperature sensors to detect hotspots improve the robustness and reliability of the system by preventing overheating. Low area and low energy consumption are essential for integrated temperature sensors in such applications. The fabricated oscillator has a ten times less footprint than state-of-the-art temperature sensing cores ( $42.3~mu {mathrm { m}}^{2} $ ) and enables low energy temperature to the digital converter (0.32 nJ energy/conversion) in GF45RFSOI technology. The proposed oscillator facilitates an area and energy-efficient temperature sensor (20 °C to 90 °C) with a simple counter-based digital readout with a best-in-class resolution figure of merit of 0.16 pJK2.
{"title":"A 42.3 μm² Band to Band Tunneling-Based Oscillator Enabled Temperature to Digital Converter With Resolution FoM of 0.16 pJK² for Embedded Temperature Sensing","authors":"Abhishek A. Kadam;Shubham Patil;Ajay K. Singh;Maryam Shojaei Baghini;Udayan Ganguly;Laxmeesha Somappa","doi":"10.1109/LSSC.2024.3433610","DOIUrl":"https://doi.org/10.1109/LSSC.2024.3433610","url":null,"abstract":"In advanced high-speed integrated systems, the widely distributed and proliferation of temperature sensors to detect hotspots improve the robustness and reliability of the system by preventing overheating. Low area and low energy consumption are essential for integrated temperature sensors in such applications. The fabricated oscillator has a ten times less footprint than state-of-the-art temperature sensing cores (\u0000<inline-formula> <tex-math>$42.3~mu {mathrm { m}}^{2} $ </tex-math></inline-formula>\u0000) and enables low energy temperature to the digital converter (0.32 nJ energy/conversion) in GF45RFSOI technology. The proposed oscillator facilitates an area and energy-efficient temperature sensor (20 °C to 90 °C) with a simple counter-based digital readout with a best-in-class resolution figure of merit of 0.16 pJK2.","PeriodicalId":13032,"journal":{"name":"IEEE Solid-State Circuits Letters","volume":null,"pages":null},"PeriodicalIF":2.2,"publicationDate":"2024-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141966120","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
High-speed links require fast, moderate resolution analog-to-digital converters (ADCs) with low power to maximize efficiency. Hybrid voltage and time (V+T) ADCs can combine the speed benefits of time-domain conversion with the reliability of conventional voltage-domain ADCs. This letter demonstrates 1) how the V+T architecture can simplify time interleaving implementation and 2) highlights two methods for improving V+T sub-ADC robustness: a) a voltage-to-time converter (VTC) with common-mode input voltage tracking and b) a merged time-to-voltage and flash time-to-digital converter. This is demonstrated in a 0.103-mm2 22-nm CMOS prototype that consumes 76 mW and gives 32.3-dB SNDR with a Nyquist input at 40 GS/s, for 57-fJ/step FoMw.
{"title":"A 7-b 76-mW 40-GS/s Hybrid Voltage/Time-Domain ADC With Common-Mode Input Tracking","authors":"Amy Whitcombe;Somnath Kundu;Hariprasad Chandrakumar;Abhishek Agrawal;Thomas Brown;Steven Callender;Brent Carlton;Stefano Pellerano","doi":"10.1109/LSSC.2024.3430851","DOIUrl":"https://doi.org/10.1109/LSSC.2024.3430851","url":null,"abstract":"High-speed links require fast, moderate resolution analog-to-digital converters (ADCs) with low power to maximize efficiency. Hybrid voltage and time (V+T) ADCs can combine the speed benefits of time-domain conversion with the reliability of conventional voltage-domain ADCs. This letter demonstrates 1) how the V+T architecture can simplify time interleaving implementation and 2) highlights two methods for improving V+T sub-ADC robustness: a) a voltage-to-time converter (VTC) with common-mode input voltage tracking and b) a merged time-to-voltage and flash time-to-digital converter. This is demonstrated in a 0.103-mm2 22-nm CMOS prototype that consumes 76 mW and gives 32.3-dB SNDR with a Nyquist input at 40 GS/s, for 57-fJ/step FoMw.","PeriodicalId":13032,"journal":{"name":"IEEE Solid-State Circuits Letters","volume":null,"pages":null},"PeriodicalIF":2.2,"publicationDate":"2024-07-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141966121","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-07-10DOI: 10.1109/LSSC.2024.3426336
Yahia Ibrahim;Ali Niknejad
This letter introduces a W-Band sequential power amplifier (PA) (Lehmann and Knoechel, 2008) with a novel output network designed to minimize passive and combiner losses, while reducing the overall footprint compared to conventional sequential and Doherty PAs (Doherty, 1936). An isolated output combiner sums two PAs operating in two different modes: 1) the main amplifier operates in class AB and 2) the auxiliary amplifier operates in class C. The measured PA achieves a saturated output power $(mathbf {P_{mathrm { sat}}})$ of 13 dBm and a gain of 12.5 dB with 3-dB bandwidth (BW) from 79.5 to 94.5 GHz. Additionally, it demonstrates a peak power-added efficiency (PAE) of 19.4% and a 14.6% PAE at 6-dB power back-off (PBO) at 87.5 GHz. Furthermore, the PA achieves a data rate of 12 Gb/s for a 16QAM signal with an average output power of 5 dBm, an average PAE of 10%, and an EVM (RMS) of -20 dB. The PA was fabricated in 16-nm FinFet technology with core area of 0.15 mm2. To the authors’ knowledge, this PA has the highest PAE at 6-dB PBO for CMOS PAs operating in the W-Band.
{"title":"A 2-Way W-Band Power Amplifier With an Isolated Combining Output Network for Power Back-Off Efficiency Enhancement in 16-nm FinFet Technology","authors":"Yahia Ibrahim;Ali Niknejad","doi":"10.1109/LSSC.2024.3426336","DOIUrl":"https://doi.org/10.1109/LSSC.2024.3426336","url":null,"abstract":"This letter introduces a W-Band sequential power amplifier (PA) (Lehmann and Knoechel, 2008) with a novel output network designed to minimize passive and combiner losses, while reducing the overall footprint compared to conventional sequential and Doherty PAs (Doherty, 1936). An isolated output combiner sums two PAs operating in two different modes: 1) the main amplifier operates in class AB and 2) the auxiliary amplifier operates in class C. The measured PA achieves a saturated output power \u0000<inline-formula> <tex-math>$(mathbf {P_{mathrm { sat}}})$ </tex-math></inline-formula>\u0000 of 13 dBm and a gain of 12.5 dB with 3-dB bandwidth (BW) from 79.5 to 94.5 GHz. Additionally, it demonstrates a peak power-added efficiency (PAE) of 19.4% and a 14.6% PAE at 6-dB power back-off (PBO) at 87.5 GHz. Furthermore, the PA achieves a data rate of 12 Gb/s for a 16QAM signal with an average output power of 5 dBm, an average PAE of 10%, and an EVM (RMS) of -20 dB. The PA was fabricated in 16-nm FinFet technology with core area of 0.15 mm2. To the authors’ knowledge, this PA has the highest PAE at 6-dB PBO for CMOS PAs operating in the W-Band.","PeriodicalId":13032,"journal":{"name":"IEEE Solid-State Circuits Letters","volume":null,"pages":null},"PeriodicalIF":2.2,"publicationDate":"2024-07-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141964880","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-06-27DOI: 10.1109/LSSC.2024.3420117
Gunmo Koo;Jaejin Kim;Seongmin Lee;Jae Hoon Shim;Kunhee Cho
An ultrawide load-range output capacitor-less digital LDO (DLDO) with an adaptive gate modulation scheme is described. The proposed DLDO is primarily regulated by digital codes with a synchronous clock signal while the gate driving level is dynamically adjusted according to the load current level. The proposed gate modulation scheme can significantly widen the dynamic range of load current and reduce the output voltage ripple. In addition, an asynchronous droop detection circuit, coupled with adaptive gate modulation, is added to improve the voltage droop and ensure fast recovery from load transients. The proposed DLDO was fabricated in 28-nm CMOS process. The dynamic load range of 57 $143times $ (1.4– $80~mu $ A) is achieved and the output voltage ripple of under 17 mV is measured across the entire load current range. A response time of less than 10 ns and a recovery time of less than 30 ns are measured in various load transient conditions.
{"title":"An Ultrawide Load-Range Fast-Transient Output Capacitor-Less Digital LDO With Adaptive Gate Modulation and Droop Detection","authors":"Gunmo Koo;Jaejin Kim;Seongmin Lee;Jae Hoon Shim;Kunhee Cho","doi":"10.1109/LSSC.2024.3420117","DOIUrl":"https://doi.org/10.1109/LSSC.2024.3420117","url":null,"abstract":"An ultrawide load-range output capacitor-less digital LDO (DLDO) with an adaptive gate modulation scheme is described. The proposed DLDO is primarily regulated by digital codes with a synchronous clock signal while the gate driving level is dynamically adjusted according to the load current level. The proposed gate modulation scheme can significantly widen the dynamic range of load current and reduce the output voltage ripple. In addition, an asynchronous droop detection circuit, coupled with adaptive gate modulation, is added to improve the voltage droop and ensure fast recovery from load transients. The proposed DLDO was fabricated in 28-nm CMOS process. The dynamic load range of 57\u0000<inline-formula> <tex-math>$143times $ </tex-math></inline-formula>\u0000 (1.4–\u0000<inline-formula> <tex-math>$80~mu $ </tex-math></inline-formula>\u0000A) is achieved and the output voltage ripple of under 17 mV is measured across the entire load current range. A response time of less than 10 ns and a recovery time of less than 30 ns are measured in various load transient conditions.","PeriodicalId":13032,"journal":{"name":"IEEE Solid-State Circuits Letters","volume":null,"pages":null},"PeriodicalIF":2.2,"publicationDate":"2024-06-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141965281","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-06-26DOI: 10.1109/LSSC.2024.3419722
Jieun Park;Yong Ki Lee;Karpinskyy Bohdan;Yunhyeok Choi;Jonghoon Shin;Hyo-Gyuem Rhew;Jongshin Shin
This letter presents a high-performance true random number generator (TRNG) based on self-timed ring (STR), showing robust tolerance to PVT variations. The evaluations were performed over 320 chips (64 chips per process corner of nn, ff, ss, sf, and fs) across three voltages (0.75 V, 0.75 V±10%) and three temperatures ( $- 40~^{circ }$ C, $25~^{circ }$ C, and $150~^{circ }$ C). All 320 test chips demonstrated stable random generation at 60 Mb/s over all the test combinations without a single failure. The verification utilized a TRNG BIST, ensuring a minimum of 0.5 min-entropy per bit. Moreover, a mathematical model for the proposed TRNG is developed to derive the throughput and the entropy of the random output.
{"title":"A PVT-Tolerant STR-Based TRNG in 4-nm Achieving 60 Mbp/s and Its Performance Analysis via Mathematical Modeling","authors":"Jieun Park;Yong Ki Lee;Karpinskyy Bohdan;Yunhyeok Choi;Jonghoon Shin;Hyo-Gyuem Rhew;Jongshin Shin","doi":"10.1109/LSSC.2024.3419722","DOIUrl":"https://doi.org/10.1109/LSSC.2024.3419722","url":null,"abstract":"This letter presents a high-performance true random number generator (TRNG) based on self-timed ring (STR), showing robust tolerance to PVT variations. The evaluations were performed over 320 chips (64 chips per process corner of nn, ff, ss, sf, and fs) across three voltages (0.75 V, 0.75 V±10%) and three temperatures (\u0000<inline-formula> <tex-math>$- 40~^{circ }$ </tex-math></inline-formula>\u0000C, \u0000<inline-formula> <tex-math>$25~^{circ }$ </tex-math></inline-formula>\u0000C, and \u0000<inline-formula> <tex-math>$150~^{circ }$ </tex-math></inline-formula>\u0000C). All 320 test chips demonstrated stable random generation at 60 Mb/s over all the test combinations without a single failure. The verification utilized a TRNG BIST, ensuring a minimum of 0.5 min-entropy per bit. Moreover, a mathematical model for the proposed TRNG is developed to derive the throughput and the entropy of the random output.","PeriodicalId":13032,"journal":{"name":"IEEE Solid-State Circuits Letters","volume":null,"pages":null},"PeriodicalIF":2.2,"publicationDate":"2024-06-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142276462","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-06-24DOI: 10.1109/LSSC.2024.3418099
Peiyu Chen;Meng Wu;Wentao Zhao;Yufei Ma;Tianyu Jia;Le Ye
In this letter, we present an analog compute-in-memory (CIM) macro design which incorporates near-CIM analog memory and nonlinearity activation unit (NAU) to alleviate the DAC/ADC power bottleneck. Fully differential analog memory is designed with switched capacitor storage circuits. Activation function, e.g., rectified linear unit, is also performed in analog domain in NAU. The CIM macro is fabricated using TSMC 55-nm technology, with a peak macro-level efficiency of 44.3 TOPS/W and a system energy efficiency of 27.7 TOPS/W for analog input and output with 4-bit weight. The near-CIM analog memory and NAU solution brings 76.0% energy reduction compared with DAC/ADC solution, which contributes $1.34times $ to $2.37times $ energy efficiency improvement.
{"title":"A 44.3 TOPS/W SRAM Compute-in-Memory With Near-CIM Analog Memory and Activation for DAC/ADC-Less Operations","authors":"Peiyu Chen;Meng Wu;Wentao Zhao;Yufei Ma;Tianyu Jia;Le Ye","doi":"10.1109/LSSC.2024.3418099","DOIUrl":"https://doi.org/10.1109/LSSC.2024.3418099","url":null,"abstract":"In this letter, we present an analog compute-in-memory (CIM) macro design which incorporates near-CIM analog memory and nonlinearity activation unit (NAU) to alleviate the DAC/ADC power bottleneck. Fully differential analog memory is designed with switched capacitor storage circuits. Activation function, e.g., rectified linear unit, is also performed in analog domain in NAU. The CIM macro is fabricated using TSMC 55-nm technology, with a peak macro-level efficiency of 44.3 TOPS/W and a system energy efficiency of 27.7 TOPS/W for analog input and output with 4-bit weight. The near-CIM analog memory and NAU solution brings 76.0% energy reduction compared with DAC/ADC solution, which contributes \u0000<inline-formula> <tex-math>$1.34times $ </tex-math></inline-formula>\u0000 to \u0000<inline-formula> <tex-math>$2.37times $ </tex-math></inline-formula>\u0000 energy efficiency improvement.","PeriodicalId":13032,"journal":{"name":"IEEE Solid-State Circuits Letters","volume":null,"pages":null},"PeriodicalIF":2.2,"publicationDate":"2024-06-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142430727","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-06-24DOI: 10.1109/LSSC.2024.3418523
Hamin Lee;Juwon Ham;Junmin Lee;Wooseok Jang;Seunghoon Ko
This letter presents a touch readout integrated circuit (IC) integrating both mutual- and self-capacitance sensing capabilities. The proposed IC aims to compensate for self-capacitance up to 620 pF by employing a combination of current-mode and capacitive-mode compensation techniques. A noise-monitoring scheme, based on the orthogonality of multicapacitance driving sequences, enhances readout performance by selectively detecting external noises during mutual-capacitance sensing operation. The IC achieved the measured signal-to-noise ratio (SNR) of 47.3, 30.6, and 36.1 dB in mutual-capacitance sensing and self-capacitance sensing of T/RX electrodes, respectively. By applying the noise-monitoring scheme, a 7-times higher noise power compared to the absence of external noise were successfully detected.
{"title":"A 620-pF-Compensated Dual-Mode Capacitance Readout IC for Subdisplay Panel Applications","authors":"Hamin Lee;Juwon Ham;Junmin Lee;Wooseok Jang;Seunghoon Ko","doi":"10.1109/LSSC.2024.3418523","DOIUrl":"https://doi.org/10.1109/LSSC.2024.3418523","url":null,"abstract":"This letter presents a touch readout integrated circuit (IC) integrating both mutual- and self-capacitance sensing capabilities. The proposed IC aims to compensate for self-capacitance up to 620 pF by employing a combination of current-mode and capacitive-mode compensation techniques. A noise-monitoring scheme, based on the orthogonality of multicapacitance driving sequences, enhances readout performance by selectively detecting external noises during mutual-capacitance sensing operation. The IC achieved the measured signal-to-noise ratio (SNR) of 47.3, 30.6, and 36.1 dB in mutual-capacitance sensing and self-capacitance sensing of T/RX electrodes, respectively. By applying the noise-monitoring scheme, a 7-times higher noise power compared to the absence of external noise were successfully detected.","PeriodicalId":13032,"journal":{"name":"IEEE Solid-State Circuits Letters","volume":null,"pages":null},"PeriodicalIF":2.2,"publicationDate":"2024-06-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141964919","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Artificial intelligence workloads demand a wide range of multiply and accumulate (MAC) precision. Pitch-matching constraints in compute-in-memory (CIM) engines limit the analog-to-digital converter (ADC) precision to about 8 bits. This letter demonstrates a method of mapping a suitable input conditioned MAC range to the input dynamic range of the on-chip 7-b ADC, thereby achieving up to 10 bits of output MAC precision. A 424 Kb SRAM CIM macro was fabricated in TSMC 28 nm, which computes 72 MACs in parallel per cycle. Measurement results at nominal supply voltage show an energy efficiency of 196.6–102 TOPS/W/b for a 2–10 bit output MAC precision. Inference results on MNIST, CIFAR10, and CIFAR100 are shown with $leq 1%$ accuracy loss from the software baseline.
{"title":"A 2-to-10-b Output Precision Reconfigurable Compute-In-Memory Macro Leveraging Input Conditioning Using Residue Amplification","authors":"Balaji Vijayakumar;Ashwin Balagopal Sundar;Janakiraman Viraraghavan;Varchas Bharadwaj","doi":"10.1109/LSSC.2024.3415476","DOIUrl":"https://doi.org/10.1109/LSSC.2024.3415476","url":null,"abstract":"Artificial intelligence workloads demand a wide range of multiply and accumulate (MAC) precision. Pitch-matching constraints in compute-in-memory (CIM) engines limit the analog-to-digital converter (ADC) precision to about 8 bits. This letter demonstrates a method of mapping a suitable input conditioned MAC range to the input dynamic range of the on-chip 7-b ADC, thereby achieving up to 10 bits of output MAC precision. A 424 Kb SRAM CIM macro was fabricated in TSMC 28 nm, which computes 72 MACs in parallel per cycle. Measurement results at nominal supply voltage show an energy efficiency of 196.6–102 TOPS/W/b for a 2–10 bit output MAC precision. Inference results on MNIST, CIFAR10, and CIFAR100 are shown with \u0000<inline-formula> <tex-math>$leq 1%$ </tex-math></inline-formula>\u0000 accuracy loss from the software baseline.","PeriodicalId":13032,"journal":{"name":"IEEE Solid-State Circuits Letters","volume":null,"pages":null},"PeriodicalIF":2.2,"publicationDate":"2024-06-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141991495","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Adapting a continuous time delta-sigma analog-to-digital converter (ADC) for incremental operation at high sampling rates degrades the noise and distortion due to potential overload of the modulator as it comes out of reset and nonlinear residue on the reset switch due to input current flowing through it in the reset phase. It is shown that the input and DAC currents must simultaneously begin to flow through the first integrating capacitor to minimize the possibility of overload. The first integrator reset has to be released just before the start of the DAC pulse. A feedforward path must be used to ensure that the DAC output is close to the input signal from the beginning. Blocking the input current from flowing through the reset switch in the reset phase eliminates the effect of the nonlinear residue. A 320 MS/s fourth-order incremental delta-sigma ADC prototype in an 180nm process using the above techniques has 90 dB dynamic range, 82 dB SNDR, and 84.5 dB SNR in a 2.5 MHz bandwidth. It consumes 46.3 mW from a 1.8V supply and occupies 0.7 mm2.
{"title":"An 81.5dB SNDR, 2.5 MHz Bandwidth Incremental Continuous-Time Delta-Sigma ADC in 180 nm CMOS","authors":"Aswani Kumar Unnam;Paramita Banerjee;Nagendra Krishnapura","doi":"10.1109/LSSC.2024.3412634","DOIUrl":"https://doi.org/10.1109/LSSC.2024.3412634","url":null,"abstract":"Adapting a continuous time delta-sigma analog-to-digital converter (ADC) for incremental operation at high sampling rates degrades the noise and distortion due to potential overload of the modulator as it comes out of reset and nonlinear residue on the reset switch due to input current flowing through it in the reset phase. It is shown that the input and DAC currents must simultaneously begin to flow through the first integrating capacitor to minimize the possibility of overload. The first integrator reset has to be released just before the start of the DAC pulse. A feedforward path must be used to ensure that the DAC output is close to the input signal from the beginning. Blocking the input current from flowing through the reset switch in the reset phase eliminates the effect of the nonlinear residue. A 320 MS/s fourth-order incremental delta-sigma ADC prototype in an 180nm process using the above techniques has 90 dB dynamic range, 82 dB SNDR, and 84.5 dB SNR in a 2.5 MHz bandwidth. It consumes 46.3 mW from a 1.8V supply and occupies 0.7 mm2.","PeriodicalId":13032,"journal":{"name":"IEEE Solid-State Circuits Letters","volume":null,"pages":null},"PeriodicalIF":2.2,"publicationDate":"2024-06-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141618102","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：